Rank: Member
Groups: Member
Joined: 2/17/2007(UTC)
Posts: 45
Location: france
|
after I mounted 2 RevC together with darwin and joshua as passive preamp into the case, I noticed there's sometimes short blob when changing volume; it's not immediately reproducible when turning forward/backward around the position where a blob occured, but it seems to happen always at the same of a few volume positions.
I don't change volume often, but I wanted to know if this behaviour is normal.
if there shouldn't be any blob, is it possible that this particular realization of an attenuator with a relay switched of a bank of discret resistors might be sensible to impedance difference between source output [with passive preamp] and RevC input? - maybe not all relays might go off or on exactly at the same time and maybe this sometimes produces a very short moment between relay states where a bigger bit difference in the binary code occurs?
|
|
|
|
|
|
Rank: Administration
Groups: Administration, Customer
Joined: 10/24/2006(UTC)
Posts: 3,979
Location: Nashville, TN
Thanks: 25 times
Was thanked: 89 time(s) in 83 post(s)
|
Hi Ullus, The controller switches the relays one at a time to reduce audible "pops" if it changed them all at once it would be faster but you would get some nasty pops almost all the time. Instead it switches them one by one. This can cause the larger steps (higher order bits) to have a slightly longer delays in switching volumes (milliseconds) this is probably what you are noticing. It is completely normal. :) Cheers! Russ Edited by user Monday, June 25, 2007 12:15:03 PM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Rank: Member
Groups: Member
Joined: 2/17/2007(UTC)
Posts: 45
Location: france
|
Hi Russ, I propose for next revision, instead of using the binary code sequence which corresponds to decimal numbers (which I suspect you use because you mention larger steps), you generate a sequence with the PIC which changes the least number of bits in each counting step and just rearrange the placement of the resistors correspondingly to the the same resistance sequence as before. a sequence which changes only one bit at each step is the best*), e.g.: for a 4 bit code, instead of a use sequence b: a b ========= 0001 0001 0010 0011 0011 0111 0100 1111 0101 1110 0110 1100 0111 1000 1000 1001 1001 1011 1010 1010 1011 0010 1100 0110 1101 0100 1110 0101 1111 1101 ========= sadly, changing the firmware only is not enough to employ the new sequence. since I don't want to de-solder so many parts, if the blobs diminish or disappear using this method, I'd like to get a free new JT for this idea which is as old as the bit itself ;) thank you in advance! I guess you don't even need to change the PCB; just make a new firmware and relabel the supplied resistors. bye ullus -- while unix sounds like a bug, C sounds like its own virus ;) this is why I can't read C; it's a write only language! *) Gray codeEdited by user Tuesday, June 26, 2007 4:28:05 AM(UTC)
| Reason: Not specified
|
|
|
|
|
|
Forum Jump
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.
